Stay updated with the latest news about AI agents, autonomous AI, and automation tools.
OpenAI's agent reached the internet from a training sandbox and ran for 2.5 hours, as US lawmakers push for AI kill switches.
OpenAI confirms its own AI agents leaked 53 ChatGPT user images to unlisted third-party links, raising new questions about agentic AI safety.
CNN’s Hadas Gold, Hilary Whiteman and Max Saltman contributed to this report.
- [Rogue OpenAI agents targeted three separate US government websites](https://www.cnn.com/2026/09/26/tech/openai-agents-rogue-government-websites?utm_source=business_ribbon) # Rogue OpenAI agents targeted three separate US government websites PUBLISHED Sep 26, 2026, 10:54 AM ET  Rogue OpenAI agents targeted US government websites, including the Commerce Department and Securities and Exchange Commission. OpenAI said Friday that some of its AI agents went rogue and probed US government websites this summer — the latest revelation of the artificial intelligence company’s technology. The New York Times first reported that the AI agents went rogue and attempted to gain access to the Education Department, the Commerce Department and the Securities and Exchange Commission, according to security researchers at AI research lab Transluce. OpenAI said Saturday that its agents accessed publicly available data from the Commerce Department’s Census Bureau using login credentials it found online, and separately shared public data from the SEC website on another website. OpenAI’s agents attempted but failed to gain access to the Education Department and gather data from its civil rights office, according to the report. OpenAI told CNN in an email that it notified the agencies of the findings while continuing an “extensive review of misaligned model activity.” Jay Obernolte, the Republican co-chair of the AI caucus, told CNN’s Anderson Cooper on Friday that the incident is “another example of a loss of human control.” The report comes just days after Australia’s prime minister said that an OpenAI agent hacked into the country’s national healthcare database, marking the [first known case](https://www.cnn.com/2026/09/23/business/australia-openai-a
SK hynix's U.S. storage subsidiary is considering a listing as early as 2027. The plan comes amid broader activity in AI infrastructure and enterprise...
# Trebellar Secures $18M Series A to Replace Spreadsheets with AI Real Estate Software ## Replacing Legacy Spreadsheets with Real-Time AI _More Update: [Palma.ai Secures $1.8M Pre-Seed Funding to Govern Enterprise AI Agents](https://ventureburn.com/palma-ai-secures-1-8m-funding)_ ## Enterprise Traction and Market Shift [Ema Raises $77 Million To Replace Enterprise Software With AI Agent Employees](https://ventureburn.com/ema-raises-77-million-scale-ai-employees-platform/) [Palma.ai Secures $1.8M Pre-Seed Funding to Govern Enterprise AI Agents](https://ventureburn.com/palma-ai-secures-1-8m-funding/) [](https://ventureburn.com/basalt-health-raises-20m-ai-post-acute-care/)
# ByteAsk raises $1 million in funding to build AI coding agents for C and C++  [ISN Team](https://indianstartupnews.com/author/isn-team-1) 26 Sep 2026 16:34 IST [ Follow Us](https://news.google.com/publications/CAAqMggKIixDQklTR3dnTWFoY0tGV2x1WkdsaGJuTjBZWEowZFhCdVpYZHpMbU52YlNnQVAB?hl=en-IN&gl=IN&ceid=IN%3Aen) [fb](http://www.facebook.com/sharer/sharer.php?u=https://indianstartupnews.com/funding/byteask-raises-1-million-in-funding-to-build-ai-coding-agents-for-c-and-c-12570766&t=ByteAsk%20raises%20$1%20million%20in%20funding%20to%20build%20AI%20coding%20agents%20for%20C%20and%20C++) [twitter](http://www.twitter.com/intent/tweet?url=https://indianstartupnews.com/funding/byteask-raises-1-million-in-funding-to-build-ai-coding-agents-for-c-and-c-12570766&text=ByteAsk%20raises%20%241%20million%20in%20funding%20to%20build%20AI%20coding%20agents%20for%20C%20and%20C%2B%2B) [whatsapp](https://api.whatsapp.com/send?text=https://indianstartupnews.com/funding/byteask-raises-1-million-in-funding-to-build-ai-coding-agents-for-c-and-c-12570766) [linkedin](https://www.linkedin.com/sharing/share-offsite/?url=https://indianstartupnews.com/funding/byteask-raises-1-million-in-funding-to-build-ai-coding-agents-for-c-and-c-12570766&title=ByteAsk%20raises%20$1%20million%20in%20funding%20to%20build%20AI%20coding%20agents%20for%20C%20and%20C++) New Update L-R Pratyush Saini Co-founder and CTO ByteAsk, Anirudha Kulkarni, co-founder and CEO ByteASK AI [coding](https://indianstartupnews.com/tags/coding) startup ByteAsk has raised $1 million in pre-seed funding from [Y Combinator](https://indianstartupnews.com/tags/Y-Combinator), Entrepreneur First and [angel investors](https://indianstartupnews.com/tags/angel-investors), as it looks to develop AI tools for engineers working with C and C++. Advertisment [ conference in Austin on September 25, Federal Trade Commission Chair Andrew Ferguson offered the clearest federal position yet on a question that consumer-facing AI products have made urgent: who pays when an autonomous agent makes a purchasing decision, books a trip, or sends an email that causes harm?
An OpenAI committee meant to have the final word on the company's safety protocols is facing scrutiny after high-profile incidents involving rogue AI...
One concrete question reveals whether the product chooses its own path or simply runs predefined instructions.
# Inside Meta's grind to turn Muse from a clunky chatbot into an AI sensation Meta also snapped up top AI agent talent through acqui-hires, including the leaders of Dreamer, an AI agent startup. ## **Nat Friedman grinds to turn Muse into a hit** ### Want more Business Insider in your news feed? When the independent [AI agent OpenClaw](https://www.businessinsider.com/openclaw-internet-viral-lobster-hats-claw-hands-poses-tech-china-2026-3) took the tech world by storm this winter, Friedman acted quickly. ## **Meta goes on an AI hiring spree** It hired David Singleton and Hugo Barra, the cofounders of [Dreamer, a startup](https://www.businessinsider.com/meta-superintelligence-labs-taps-leader-for-hardware-role-2026-4) helping people build their own agents, along with the rest of their team. Although the arrival of the viral AI agent [Instinct initially upstaged Meta](https://www.businessinsider.com/noah-shinn-instinct-ai-competes-meta-muse-2026-9), Muse took off by tapping into Meta's enormous built-in audience across its apps — including among people who had never used AI agents before.
OpenAI disclosed Friday that its artificial intelligence agents had interacted with several U.S. government websites in unexpected ways, as part of an ongoing review into the company's models' unanticipated behavior. AI evaluator and research lab Transluce said Friday that through an independent investigation it also found that agents appearing to originate from OpenAI attempted a rudimentary hack on a Department of Education website for the department's civil rights office, which did not succeed. #### CBS News 24/7 CBS News App[Open](https://www.cbsnews.com/app/?ftag=CNM-16-10abd6g&url=https://www.cbsnews.com/news/openai-ai-agent-bot-rogue-hack-government-website/)
The first documented campaign where autonomous agents executed the full breach lifecycle at scale — vulnerability discovery, exploitation, exfiltration,...
CrowdStrike Holdings (NasdaqGS:CRWD) has joined peers as a founding member of the new Blueprint Alliance to secure AI agents. The Blueprint Alliance brings...
# Stop watching what AI agents say and start watching what they do ##### Tell us about a guardrail that was too aggressive and broke legitimate agent workflows. How did you tune it, and what tradedoff did you land on between safety and usefulness? It is safe, but it also stops agents from doing basic investigation and makes them nearly useless during incidents. If a human has to approve every read or routine diagnostic step, much of the value of autonomy disappears. Read-only queries within an approved scope can run automatically. Reversible changes receive tighter checks. Irreversible or high-blast-radius actions require human approval. That also means looking at what an action can actually change, rather than treating every interaction with a production environment as equally dangerous. The tradeoff is not safety vs usefulness. It is about applying friction in proportion to potential impact. ## **Featured** news - [OpenAI agent hacking spree widens to Australia, targeting government website](https://www.helpnetsecurity.com/2026/09/24/openai-agent-hacking-australia/) - [New Android malware RemControl steals banking PINs and blocks removal attempts](https://www.helpnetsecurity.com/2026/09/24/remcontrol-android-banking-trojan-fake-tv-app/) - [What to do first when you get 90 days to secure AI agent data](https://www.helpnetsecurity.com/2026/09/24/kelly-herrell-nol8-ai-agent-data-security/) ## **Don't** miss - [OpenAI agent hacking spree widens to Australia, targeting government website](https://www.helpnetsecurity.com/2026/09/24/openai-agent-hacking-australia/) - [New Android malware RemControl steals banking PINs and blocks removal attempts](https://www.helpnetsecurity.com/2026/09/24/remcontrol-android-banking-trojan-fake-tv-app/) - [DarkMe RAT trades zero-days for plain phishing emails](https://www.helpnetsecurity.com/2026/09/23/darkme-rat-phishing-email-hits-corporate-targets/) Cybersecurity news Periodical newsletter released for important security events
of [Sierra](https://sierra.ai/), which builds AI agents for other companies, outlined his argument ## Safety conversation e-commerce remains uncertain, following Amazon’s decision to ban Muse, [Meta's new personal AI agent](https://www.phocuswire.com/news/technology/meta-launches-ai-agent-travel-booking), from shopping on ## From Our Partners [](https://www.phocuswire.com/news/online/expedia-hotel-booking-meta-muse) [Expedia to offer hotel bookings through Meta's Muse AI agent](https://www.phocuswire.com/news/online/expedia-hotel-booking-meta-muse) [](https://www.phocuswire.com/news/technology/openai-chairman-bret-taylor-tells-travel-industry-do-not-restrain-ai-agents) [OpenAI chairman tells travel industry: Do not restrain your AI agents](https://www.phocuswire.com/news/technology/openai-chairman-bret-taylor-tells-travel-industry-do-not-restrain-ai-agents) [](https://www.phocuswire.com/news/online/expedia-hotel-booking-meta-muse) [Expedia to offer hotel bookings through Meta's Muse AI agent](https://www.phocuswire.com/news/online/expedia-hotel-booking-meta-muse)
AI agents are learning to evade rules, hide information, and resist oversight — just like the Silicon Valley titans who built them.
AxLabs has released Simple Agent Wallet, a command-line wallet designed for AI agents to autonomously pay for resources with Neo X supported.
# Businesses Focus Areas For AI And AI Agents Deployment ## AI and AI Agents Business Focus Areas: Final Thoughts Premium Journalism, deeply reported stories and breaking news
[Skip to content](https://www.krqe.com/news/albuquerque-metro/report-open-ai-ai-agents-tried-to-hack-into-multiple-systems-including-unm-digital-library/#content) According to a study from the nonprofit Transluce, AI agents from OpenAI attempted to breach UNM’s digital library on May 25 and 26 to retrieve a photo from its Valmora collection. Transluce said agents also tried to hack into a data visualization firm, Data USA, as well as the Australian Institute of Health and Welfare. [Read next >](https://www.krqe.com/news/albuquerque-metro/report-open-ai-ai-agents-tried-to-hack-into-multiple-systems-including-unm-digital-library/)  [Read next >](https://www.krqe.com/news/albuquerque-metro/report-open-ai-ai-agents-tried-to-hack-into-multiple-systems-including-unm-digital-library/) [Next](https://www.krqe.com/news/albuquerque-metro/report-open-ai-ai-agents-tried-to-hack-into-multiple-systems-including-unm-digital-library/) [Read next >](https://www.krqe.com/news/albuquerque-metro/report-open-ai-ai-agents-tried-to-hack-into-multiple-systems-including-unm-digital-library/)  [Next story in](https://www.krqe.com/news/albuquerque-metro/report-open-ai-ai-agents-tried-to-hack-into-multiple-systems-including-unm-digital-library/) [Read next >](https://www.krqe.com/news/albuquerque-metro/report-open-ai-ai-agents-tried-to-hack-into-multiple-systems-including-unm-digital-library/) [Next story in](https://www.krqe.com/news/albuquerque-metro/report-open-ai-ai-agents-tried-to-hack-into-multiple-systems-including-unm-digital-library/)
 [AIONA](https://www.aiona.co.jp/en), the developer of the AI agent “REVY” that automates and supports design tasks in the manufacturing industry, announced on the 24th that it has raised a total of ¥140 million. The funding was announced through a [press release](https://prtimes.jp/main/html/rd/p/000000020.000155874.html). The funding consists of the issuance of J-KISS type new stock acquisition rights through a third-party allocation and loans from Aichi Bank and Mizuho Bank. ## Post navigation ### [Manufacturing Site Improvement DX App "GEMBA" Secures ¥45 Million in Seed Funding from ANOBAKA and Others](https://thebridge.jp/en/2026/09/manufacturing-site-improvement-dx-app-gemba-secures-%c2%a545-million-in-seed-funding-from-anobaka-and-others) [back to top](https://thebridge.jp/en/2026/09/aiona-raises-%C2%A5140-million-led-by-onboard-for-ai-agent-revy-automating-design-tasks)
# Every CX start-up funding deal you need to know: September 25 ## **Kastle raises $24m in Series A led by Insight Partners** The two-year-old San Francisco start-up is building the AI workforce for financial services, beginning with consumer lending, and to date, its [AI agents](https://www.cxnetwork.com/guides/ai-agents-for-customer-experience-a-complete-guide) have processed more than $1.8 billion in transactions. Through its agents, Kastle proposes a new operating model for financial services: hybrid teams in which AI agents handle high-volume, repeatable work, while people focus on complex cases requiring judgment, expertise, empathy and relationships. Kastle deploys AI agents that can navigate complexity, pass the bar on regulatory rigor and get high-stakes work done without waiting for a multiyear transformation. ## **Baselayer raises $35m from M13 in Series A** New York-based start-up Baselayer, which provides identity verification and fraud-risk assessments for financial institutions, including credentials that verify AI agents, closed its Series A round with $35 million. The round marks the first dedicated capital injection into Know Your Agent identity infrastructure, which is the layer that lets banks, merchants and platforms verify which AI agent they are dealing with, who it represents and whether it is authorized to act. The funding round news broke within hours of [six global banks publishing](https://www.cxnetwork.com/cx-experience/news/6-global-banks-collaborate-on-principles-for-trusted-agentic-commerce) [a framework](https://www.cxnetwork.com/cx-experience/news/6-global-banks-collaborate-on-principles-for-trusted-agentic-commerce) for secure and trustworthy agent payments. ## **Sela raises $21m to mimic top sales performers for voice automated mortgage calls** Sela is developing AI voice agents that automate mortgage sales calls and connect borrowers with loan officers when required. Its AI agents help loan officers originate more than $1 bi
AI ### 4. AI agents emerge as insider risks concentrated on enterprise endpoints. Don’t miss the [full interview](https://www.youtube.com/watch?v=lhhsfqWsvMw) on theCUBE.
The prospect of legal accountability around AI is unclear, but what happens when the hackers aren't human?
# Senate Panel Will Delve Into Rogue AI Threat ## Why it Matters Autonomous AI agents broke out of their controlled environment in July 2026, compromising systems at both OpenAI and Hugging Face, putting Congress on alert about whether the federal government has any tools to respond when AI systems act without authorization.
Instinct saved me $550, booked my restaurant reservations, and warned me about a phishing scam. It also wasted $64 and might be a security nightmare.
With the $400 million in new funds announced on Thursday, the US-Israeli firm plans to hire more engineers for its Tel Aviv research and development center and expand its browser platform to help businesses manage security threats as AI agents are increasingly integrated into the everyday workplace. The Dallas-based firm was founded in 2020 by Dan Amiga, a serial entrepreneur and veteran of the Israeli army’s 8200 unit, and Mike Fey, former Symantec President. - [AI agents](https://www.timesofisrael.com/topic/ai-agents/)
In Brief: In this ITB Berlin interview, Bhanu Chopra explains how AI agents are reshaping the sales chain, becoming new gatekeepers, and why real-time data...
### Palo Alto Networks tackles the risks of autonomous AI agents Palo Alto Networks Inc. is addressing that change through [Prisma AIRS integrations](https://www.paloaltonetworks.com/sase/prisma-airs-contact-us?utm_source=google-jg-amer-portfolio-brnd-port&utm_medium=paid_search&utm_campaign=google-portfolio-ai_netsec-amer-multi-discovery-en-brand&utm_content=701Ki000000osCfIAI&utm_term=palo%20alto%20prisma%20airs&cq_plac=&cq_net=g&gclsrc=aw.ds&gad_source=1&gad_campaignid=23588641131&gbraid=0AAAAADHVeKlh6TyvXixH5rNC9tdKmEM_Q&gclid=CjwKCAjwj7HTBhBiEiwA8s35OlFMkHGqkQMzuzaaEOInTSt5RCDlzhynyi4fAy0fAlpRNMOhIFpQlhoCa_gQAvD_BwE) designed to inspect agent activity within Google Cloud environments. “We can’t treat agents like a chatbot,” Thellmann said. Thellmann spoke with theCUBE’s [John Furrier](https://www.linkedin.com/in/furrier/) for the [Google Cloud: AI Agents in Action Series](https://www.thecube.net/events/google/ai-agents-in-action) on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how Palo Alto Networks is [securing autonomous agents](https://siliconangle.com/2026/05/12/idira-launches-palo-alto-networks-extends-cyberark-tech-machine-agentic-identities/), model activity and tool interactions within Google Cloud. ### AI agent security moves beyond prompts “You’ll have these autonomous beings, non-humans, within your network that are just doing things all the time, either in isolation or together,” he added.
24, 2026 /PRNewswire/ -- As AI agents move beyond experimentation, companies must decide not just where to deploy them, but how work itself should change.
Our team at TinyAI.Tools builds bespoke AI solutions tailored to your business.